by Dr. Leslie Kanthan, CEO of TurinTech
According to a recent Business Insider report, banks can potentially save an estimated $447 billion in costs by 2023 from implementing AI applications.
It’s therefore no surprise that global AI spending is estimated to reach $110 billion by 2024. AI technology adoption is expected to grow in areas such as asset management, algorithmic trading, credit underwriting, and blockchain-based financial services.
With these figures in mind, it is clear that adopting AI is more than just a nice to have but rather something that can deliver real valuable benefits to the business.
Here are our top three factors financial leaders should consider when bringing AI technology into their financial services businesses:
- Assess AI maturity and decide which approach is right for your business – Buy, Build or Buy first build later
As we all know, many banks are still running on legacy systems which makes the deployment of newer technologies harder said than done. When it comes to adopting AI in particular, firms need to consider which approach fits them better: should they buy these capabilities and build them in-house, or buy them first so as to have at least a head start and build after?
The buy approach works well when you need AI models to automate certain back or middle office tasks and service providers are widely available in the market. Not only can this be done quickly and help save money, but they also have highly experienced talent and the right resources immediately available to deliver a successful project.
One may consider the build approach when a use case is so new, and the first of its kind, that a new AI solution needs to be developed especially for it. Investing in resources and people can therefore have benefits, including owning the code and IP of a new solution.
Lastly, if you are at the beginning of your AI transformational journey, the buy-then-build approach may be the best approach for your business. This has the advantage of requiring a smaller initial investment but still being a good starting point.
Having said this, we think that ultimately if you want to address unique use cases and leverage AI as a key differentiator, building AI in house is your best option.
- Transparency & Explainability
Financial services are embracing AI to assist with various aspects of their services including optimising investment portfolios and predicting liquidity balances.
An important factor business leaders should consider when adopting AI, however, is how they plan to solve the transparency challenges that AI brings with it. Transparency is the biggest setback for AI technology, along with the potential for bias. For example, issues with the iPhone’s facial recognition technology which can also be used as a form of a payment transaction, caused a big stir in China, as the AI technology was not able to distinguish between Asian facial features. It is this bias that calls for regulatory and governance frameworks if one is to avoid discrimination being built into AI models.
The above example demonstrates that in order to achieve the full potential of AI solutions, businesses must place more focus on increasing AI transparency. Explainable AI focuses on the rationale behind AI decision-making, helping to create more clarity and understanding for stakeholders, investors, and clients. Gaining full ownership of the code and AI model can help have ultimate transparency, which is of utmost importance as per dictated regulations and compliance.
- Choose an AI tool with scalability for the long run
One of the lessons learned from the pandemic is that a technology patchwork never works. It may provide a quick temporary fix, and offer short-term relief and benefits, but in the long term, it will be another technology trying to run on top of legacy systems. Even during the COVID-19 pandemic, when most industries plummeted, financial investments toward AI remained high. Investment in AI is breaking records and is only increasing by the year. According to CB insights, the sector had $38 billion in new investment in 2021, showcasing how financial leaders are recognising the value and long-term benefits of scaleable AI.
Leaders need to invest in solutions that not only work for now but also in the long run. This means:
having the ownership of model and code for customisability and competitive edge and being able to continuously monitor and optimise the performance of various models on a unified platform.
The bottom line
The future of AI is certainly getting brighter, as it moves away from being just a buzzword, to finally earning its place in the technology ‘Hall of Fame’. There are many benefits for finance leaders to adopt AI technology. All they need is to do their research and make sure they make an informed decision on the best approach to do so.
